• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

NV20 performance benchmarks...

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Adam:

You said that I was incorrect. How can I be incorrect in saying that nVidias 2x2 is 2 samples from horizontal and 2 samples from verticle?

uh, it's not. 2x2 means you double the resolution in the vertical and horizontal (2x vertical, 2x horizontal), and then you downsize it and "average" the pixel values.

in other words, 640x480 w/2x2FSAA means that the card takes the 640x480 and doubles each resolution to 1280x1024. Then it "scrunches" it back down to 640x480, and takes the average pixel value while downsizing (loosely explained)

so it doesn't just take 2 samples from the vertical and 2 from the horizontal.

so...you were wrong. Nothing wrong with that. we're all wrong now and again

***EDIT***

except Ben. Ben is never wrong. 😉

***EDIT***

Thats what it is, but then again I'm talking about a feature that nVidia has, something which you'll never agree with.

not really sure what you mean by that. I suppose you're trying to speak of some sort of bias? Not here. I just like the 5500 better than the GTS after using 4 of them (2 each). This is actually the first time in 3 generations I've chosen a 3dfx card over an nvidia card. Kinda weird. 😉
 
Yes, I know that supersampling requires the doubling of the screen resolution. Perhaps I am defining supersamling wrong. I've always viewed supersampling as a way in reducing aliasing and subpixel occlusion errors by computing multiple sample points within a pixel, and assigning the appropriately weighted average of the subpixel values to the pixel. In the case of 2x2 supersampling, when the supersampled image is reduced it takes the average of the two nearest pixel samples in the horizontal axis and the two nearest pixel samples in the verticle axis to produce the one pixel in the final image.
 
By the way, Soccerman question was how does nVidia's method compare to 3dfx's mothod. I was trying to state how it is done in terms of the number of pixel samples used. Supersampling is just another way of obtaining a set of pixel samples.
 
If your questioning how I should have stated where nVidia obtains the pixel samples, then yes. I should have explained that a little more.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top